Predicting stress in first-year college students using sleep data from wearable devices

Area of Science:

• Digital Health
• Wearable Technology
• Mental Health Monitoring

Background:

• Consumer wearables effectively measure sleep and physiological data.
• Quantifying the link between sleep, physiological stress, and perceived stress is challenging.
• College students face significant stress, impacting mental health.

Purpose of the Study:

• To analyze the association between sleep metrics and perceived stress in college students.
• To determine if wearable-derived physiological data can predict stress levels.
• To understand the role of sleep in managing student stress.

Main Methods:

• Weekly biometric data (sleep, heart rate, HRV, respiratory rate) and perceived stress surveys collected from 525 university students.
• Mixed-effects regression models used to analyze associations.
• Controlled for gender and week of the semester.

Main Results:

• Significant associations found between perceived stress and total sleep time (TST), resting heart rate (RHR), heart rate variability (HRV), and average respiratory rate (ARR).
• Increased TST and HRV correlated with decreased stress odds.
• Increased RHR and ARR correlated with increased stress odds.
• Gender and week of semester were significant predictors of stress.

Conclusions:

• Wearable-derived sleep and physiological data show promise for understanding and predicting student stress.
• Findings support the use of wearables in monitoring mental well-being among students.
• Interventions targeting sleep may help mitigate stress in this population.